classifiersplitevaluator.java

Java 编写的多种数据挖掘算法 包括聚类、分类、预处理等

      if (m_Template instanceof AdditionalMeasureProducer) {
	Enumeration en = ((AdditionalMeasureProducer)m_Template).
	  enumerateMeasures();
	while (en.hasMoreElements()) {
	  String mname = (String)en.nextElement();
	  for (int j=0;j<m_AdditionalMeasures.length;j++) {
	    if (mname.compareToIgnoreCase(m_AdditionalMeasures[j]) == 0) {
	      m_doesProduce[j] = true;
	    }
	  }
	}
      }
    } else {
      m_doesProduce = null;
    }
  }

  /**
   * Returns an enumeration of any additional measure names that might be
   * in the classifier
   * @return an enumeration of the measure names
   */
  public Enumeration enumerateMeasures() {
    Vector newVector = new Vector();
    if (m_Template instanceof AdditionalMeasureProducer) {
      Enumeration en = ((AdditionalMeasureProducer)m_Template).
	enumerateMeasures();
      while (en.hasMoreElements()) {
	String mname = (String)en.nextElement();
	newVector.addElement(mname);
      }
    }
    return newVector.elements();
  }
  
  /**
   * Returns the value of the named measure
   * @param additionalMeasureName the name of the measure to query for its value
   * @return the value of the named measure
   * @throws IllegalArgumentException if the named measure is not supported
   */
  public double getMeasure(String additionalMeasureName) {
    if (m_Template instanceof AdditionalMeasureProducer) {
      if (m_Classifier == null) {
	throw new IllegalArgumentException("ClassifierSplitEvaluator: " +
					   "Can't return result for measure, " +
					   "classifier has not been built yet.");
      }
      return ((AdditionalMeasureProducer)m_Classifier).
	getMeasure(additionalMeasureName);
    } else {
      throw new IllegalArgumentException("ClassifierSplitEvaluator: "
			  +"Can't return value for : "+additionalMeasureName
			  +". "+m_Template.getClass().getName()+" "
			  +"is not an AdditionalMeasureProducer");
    }
  }

  /**
   * Gets the data types of each of the key columns produced for a single run.
   * The number of key fields must be constant
   * for a given SplitEvaluator.
   *
   * @return an array containing objects of the type of each key column. The 
   * objects should be Strings, or Doubles.
   */
  public Object [] getKeyTypes() {

    Object [] keyTypes = new Object[KEY_SIZE];
    keyTypes[0] = "";
    keyTypes[1] = "";
    keyTypes[2] = "";
    return keyTypes;
  }

  /**
   * Gets the names of each of the key columns produced for a single run.
   * The number of key fields must be constant
   * for a given SplitEvaluator.
   *
   * @return an array containing the name of each key column
   */
  public String [] getKeyNames() {

    String [] keyNames = new String[KEY_SIZE];
    keyNames[0] = "Scheme";
    keyNames[1] = "Scheme_options";
    keyNames[2] = "Scheme_version_ID";
    return keyNames;
  }

  /**
   * Gets the key describing the current SplitEvaluator. For example
   * This may contain the name of the classifier used for classifier
   * predictive evaluation. The number of key fields must be constant
   * for a given SplitEvaluator.
   *
   * @return an array of objects containing the key.
   */
  public Object [] getKey(){

    Object [] key = new Object[KEY_SIZE];
    key[0] = m_Template.getClass().getName();
    key[1] = m_ClassifierOptions;
    key[2] = m_ClassifierVersion;
    return key;
  }

  /**
   * Gets the data types of each of the result columns produced for a 
   * single run. The number of result fields must be constant
   * for a given SplitEvaluator.
   *
   * @return an array containing objects of the type of each result column. 
   * The objects should be Strings, or Doubles.
   */
  public Object [] getResultTypes() {
    int addm = (m_AdditionalMeasures != null) 
      ? m_AdditionalMeasures.length 
      : 0;
    int overall_length = RESULT_SIZE+addm;
    overall_length += NUM_IR_STATISTICS;
    if (getAttributeID() >= 0) overall_length += 1;
    if (getPredTargetColumn()) overall_length += 2;
    Object [] resultTypes = new Object[overall_length];
    Double doub = new Double(0);
    int current = 0;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;

    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;

    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;

    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;

    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;

    // IR stats
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;

    // Timing stats
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;
    resultTypes[current++] = doub;

    // ID/Targets/Predictions
    if (getAttributeID() >= 0) resultTypes[current++] = "";
    if (getPredTargetColumn()){
        resultTypes[current++] = "";
        resultTypes[current++] = "";
    }
    
    // Classifier defined extras
    resultTypes[current++] = "";

    // add any additional measures
    for (int i=0;i<addm;i++) {
      resultTypes[current++] = doub;
    }
    if (current != overall_length) {
      throw new Error("ResultTypes didn't fit RESULT_SIZE");
    }
    return resultTypes;
  }

  /**
   * Gets the names of each of the result columns produced for a single run.
   * The number of result fields must be constant
   * for a given SplitEvaluator.
   *
   * @return an array containing the name of each result column
   */
  public String [] getResultNames() {
    int addm = (m_AdditionalMeasures != null) 
      ? m_AdditionalMeasures.length 
      : 0;
    int overall_length = RESULT_SIZE+addm;
    overall_length += NUM_IR_STATISTICS;
    if (getAttributeID() >= 0) overall_length += 1;
    if (getPredTargetColumn()) overall_length += 2;

    String [] resultNames = new String[overall_length];
    int current = 0;
    resultNames[current++] = "Number_of_training_instances";
    resultNames[current++] = "Number_of_testing_instances";

    // Basic performance stats - right vs wrong
    resultNames[current++] = "Number_correct";
    resultNames[current++] = "Number_incorrect";
    resultNames[current++] = "Number_unclassified";
    resultNames[current++] = "Percent_correct";
    resultNames[current++] = "Percent_incorrect";
    resultNames[current++] = "Percent_unclassified";
    resultNames[current++] = "Kappa_statistic";

    // Sensitive stats - certainty of predictions
    resultNames[current++] = "Mean_absolute_error";
    resultNames[current++] = "Root_mean_squared_error";
    resultNames[current++] = "Relative_absolute_error";
    resultNames[current++] = "Root_relative_squared_error";

    // SF stats
    resultNames[current++] = "SF_prior_entropy";
    resultNames[current++] = "SF_scheme_entropy";
    resultNames[current++] = "SF_entropy_gain";
    resultNames[current++] = "SF_mean_prior_entropy";
    resultNames[current++] = "SF_mean_scheme_entropy";
    resultNames[current++] = "SF_mean_entropy_gain";

    // K&B stats
    resultNames[current++] = "KB_information";
    resultNames[current++] = "KB_mean_information";
    resultNames[current++] = "KB_relative_information";

    // IR stats
    resultNames[current++] = "True_positive_rate";
    resultNames[current++] = "Num_true_positives";
    resultNames[current++] = "False_positive_rate";
    resultNames[current++] = "Num_false_positives";
    resultNames[current++] = "True_negative_rate";
    resultNames[current++] = "Num_true_negatives";
    resultNames[current++] = "False_negative_rate";
    resultNames[current++] = "Num_false_negatives";
    resultNames[current++] = "IR_precision";
    resultNames[current++] = "IR_recall";
    resultNames[current++] = "F_measure";
    resultNames[current++] = "Area_under_ROC";

    // Timing stats
    resultNames[current++] = "Elapsed_Time_training";
    resultNames[current++] = "Elapsed_Time_testing";
    resultNames[current++] = "UserCPU_Time_training";
    resultNames[current++] = "UserCPU_Time_testing";

    // ID/Targets/Predictions
    if (getAttributeID() >= 0) resultNames[current++] = "Instance_ID";
    if (getPredTargetColumn()){
        resultNames[current++] = "Targets";
        resultNames[current++] = "Predictions";
    }
    
    // Classifier defined extras
    resultNames[current++] = "Summary";
    // add any additional measures
    for (int i=0;i<addm;i++) {
      resultNames[current++] = m_AdditionalMeasures[i];
    }
    if (current != overall_length) {
      throw new Error("ResultNames didn't fit RESULT_SIZE");
    }
    return resultNames;
  }

  /**
   * Gets the results for the supplied train and test datasets. Now performs
   * a deep copy of the classifier before it is built and evaluated (just in case
   * the classifier is not initialized properly in buildClassifier()).
   *
   * @param train the training Instances.
   * @param test the testing Instances.
   * @return the results stored in an array. The objects stored in
   * the array may be Strings, Doubles, or null (for the missing value).
   * @throws Exception if a problem occurs while getting the results
   */
  public Object [] getResult(Instances train, Instances test)
  throws Exception {
    
    if (train.classAttribute().type() != Attribute.NOMINAL) {
      throw new Exception("Class attribute is not nominal!");
    }
    if (m_Template == null) {
      throw new Exception("No classifier has been specified");
    }
    int addm = (m_AdditionalMeasures != null) ? m_AdditionalMeasures.length : 0;
    int overall_length = RESULT_SIZE+addm;
    overall_length += NUM_IR_STATISTICS;
    if (getAttributeID() >= 0) overall_length += 1;
    if (getPredTargetColumn()) overall_length += 2;
    
    ThreadMXBean thMonitor = ManagementFactory.getThreadMXBean();
    boolean canMeasureCPUTime = thMonitor.isThreadCpuTimeSupported();
    if(!thMonitor.isThreadCpuTimeEnabled())
      thMonitor.setThreadCpuTimeEnabled(true);
    
    Object [] result = new Object[overall_length];
    Evaluation eval = new Evaluation(train);
    m_Classifier = Classifier.makeCopy(m_Template);
    double [] predictions;
    long thID = Thread.currentThread().getId();
    long CPUStartTime=-1, trainCPUTimeElapsed=-1, testCPUTimeElapsed=-1,
         trainTimeStart, trainTimeElapsed, testTimeStart, testTimeElapsed;    

    //training classifier
    trainTimeStart = System.currentTimeMillis();
    if(canMeasureCPUTime)
      CPUStartTime = thMonitor.getThreadUserTime(thID);
    m_Classifier.buildClassifier(train);    
    if(canMeasureCPUTime)
      trainCPUTimeElapsed = thMonitor.getThreadUserTime(thID) - CPUStartTime;
    trainTimeElapsed = System.currentTimeMillis() - trainTimeStart;
    
    //testing classifier